The role of medical comorbidities in rotator cuff retear after arthroscopic repair: a TriNetX database review of 98,844 patients

Abstract

Background

Arthroscopic rotator cuff repair (ARCR) is the most common form of operative intervention utilized in the treatment of rotator cuff tears. Retear rates following ARCR have been reported to range from 7% to 94%. This study was undertaken to better understand the effects of various medical comorbidities and age on rotator cuff retear risk following ARCR.

Methods

The TriNetX US Collaborative Network was queried for all patients treated with ARCR from 2013 to 2022 with recurrence of rotator cuff tear within 1 year. Cohorts were defined by inclusion of medical comorbidities at time of ARCR including hypertension, hyperlipidemia, chronic ischemic heart disease, type 2 diabetes mellitus, smoking history, and chronic kidney disease. Patient age groups in 10-year increments were also examined. Propensity score matching was performed for comorbid groups to adjust for age and comorbidities. Relative risk was calculated for each comorbidity and age group.

Results

There were a total of 98,844 patients who underwent ARCR for complete rotator cuff tear from 2012 to 2023 with an overall retear percentage of 16.3% within 1 year of surgery. Hypertension, hyperlipidemia, and smoking history were significant risk factors of retear after propensity score matching. Chronic ischemic heart disease, chronic kidney disease, and type 2 diabetes mellitus were not significant risk factors of retear after propensity score matching. It was also noted that relative risk of retear increased as patient age increased; patients aged 81-90 having the highest likelihood of retear (relative risk: 1.206), with a retear rate of 19.6%.

Conclusion

Medical comorbidities and advanced age can increase the risk of rotator cuff retear following ARCR. Understanding how these common comorbidities affect retear rate following ARCR can better inform orthopedic surgeons of potential risks and optimize postoperative outcomes.

Each year, hundreds of thousands of Americans undergo surgery to repair a torn rotator cuff. Arthroscopic rotator cuff repair (ARCR) is the most common form of operative intervention utilized in these patients with the goal of reattaching the torn tendon to its insertion site in the humeral head. Despite exceptional clinical outcomes, postoperative rotator cuff retears are common. Retears are most likely to occur within the first 3 to 6 months after the index surgery, with a wide range of reported occurrences, ranging between 7.2 and 94%.^23,31 Despite technological advances and improvements in repair techniques, the retear rates remain high. The etiology of recurrent rotator cuff tears following ARCR is multifactorial; preoperative characteristics including tear characteristics and medical comorbidities, intraoperative management of the repair, biomechanical factors, and postoperative rehabilitation strategies all play a potential role in contributing to retears in ARCR.^23,25 Following a retear after ARCR, patients are known to do significantly worse with regard to their satisfaction, pain scores, functional outcomes, and are at significantly increased risk of the development of osteoarthritis.⁴,¹⁸ Hence, better understanding of retears can improve surgical outcomes.

While revision ARCR is an option in the management of these patients, understanding preoperative, intraoperative, and postoperative risk factors for retears is paramount in mitigating recurrent rotator cuff tears. Moreover, revision ARCR surgery is technically more challenging, less predictable, and yields less reliable results when compared to primary rotator cuff repairs.¹⁵ Given the mixed literature on the impact of certain medical comorbidities and patient factors on the risk of retear, further research is needed to further determine potential relationships. As such, the purpose of this study is to examine medical comorbidity risk factors and their impact on rotator cuff retears after ARCR to optimize outcomes and improve treatment.

Methods

A retrospective study was conducted using the TriNetX US Collaborative Network, accessed on March 1, 2024. TriNetX is a widely utilized global research network that provides access to de-identified aggregate patient data, including information on procedures, diagnoses, medications, vitals, genomics, and demographics15. The network comprises data contributed by multiple Healthcare Organizations globally, which submit data in compliance with local privacy regulations using de-identified, pseudo-anonymized, or limited data sets.

The de-identification process adheres strictly to HIPAA Privacy Rule standards, ensuring that patient information is anonymized and privacy is safeguarded. This process has been verified by a qualified expert under the HIPAA de-identification requirements specified in Section §164.514 (b).¹⁵ Due to the database’s de-identified nature, it is exempt from Institutional Review Board review and approval.

TriNetX is employed extensively in clinical research, providing a broad range of real-world data used to derive insights and support evidence-based medicine. The database undergoes continuous quality checks to maintain the accuracy and reliability of its data, making it a valuable resource for retrospective analyses such as ours. By utilizing this comprehensive data source, our study aims to minimize bias and enhance the generalizability of findings across diverse patient populations.

Patients undergoing ARCR for complete rotator cuff tear, defined by inclusion of the associated Current Procedural Terminology system code (Current Procedural Terminology: 29827), between January 1, 2012 and December 31, 2021 were included, resulting in a total of 98.844 patients. Patient cohorts were subsequently defined by inclusion and exclusion of International Classification for Disease, 10th Edition diagnosis codes for included medical conditions present at time of operation, in addition to age cohorts categorized by decade intervals. (Table I). Medical comorbidities accounted for in the analysis include hypertension (HTN), hyperlipidemia (HLD), chronic ischemic heart failure (CIHD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM). Retear of the rotator cuff was determined based on International Classification for Disease, 10th Edition code of ipsilateral rotator cuff tear within 1 year of the index operation.

Table I.

ICD-10 and CPT codes utilized in TriNetX query.

Codes
ICD-10
Rotator cuff tear
Complete rotator cuff tear	Right- M75.121 Left- M75.122
Morbidity
Hypertension	I10-I1A
Hyperlipidemia	E78.5
Chronic ischemic heart disease	I25
Diabetes (T2D)	E11
Smoking	F17.21
Chronic kidney disease	N18
CPT
Arthroscopic rotator cuff repair	29827

ICD-10, International Classification for Disease, 10th Edition; CPT, Current Procedural Terminology; T2D, type 2 diabetes.

Statistical analysis consisted of relative risk (RR) calculations with 95% confidence intervals (CIs) between cohorts with diagnosis of individual medical conditions to respective medical condition free cohorts. Similar calculations were performed across age groups. The TriNetX platform was used to run 1:1 propensity score matching using logistic regression for morbid groups.³³ The platform uses nearest-neighbor matching with a tolerance level of 0.01 and difference between propensity scores ≤ 0.1. Age, gender, race, and morbidities mentioned in Table I were included in propensity score matching. For all analyses, a two tailed P value ≤ .05 was considered significant.

Results

Risk of retear by age

A total of 98,844 patients underwent ARCR for complete rotator cuff tear from 2012 to 2023 with an overall retear percentage of 16.3% within 1 year of surgery. (Table II) There was a near linear increase in risk of retear with increase in age, with patients aged 81-90 with highest likelihood of retear (RR 1.206, P < .0001 (95% CI 1.129, 1.289) with a retear percentage of 19.6%. (Fig. 1) Patients aged 61-70 had the second highest rate of retear (17.8%) with a significant increase in relative risk (RR 1.154, P < .0001 (95% CI 1.121, 1.187). (Fig. 2) Patients aged 71-80 had a retear rate of 16.8% and a significant increase in risk of retear (RR 1.034, P = .045 (95% CI 1.001, 1.069). Patients aged 51-60 had an insignificant increase in risk of retear (RR 1.003, P < .849 (95% CI 0.971, 1.037) with a retear rate of 16.4%.

Table II.

Retear percentage and relative risk by age group.

	RCR (2012-2021)	Retear	Retear percentage	RR	95% CI	P value
Total	97125	16146	16.62%
Age
81-90	3749	735	19.61%	1.206	1.1290, 1.289	<.0001
71-80	22853	3841	16.81%	1.034	1.001, 1.069	.0452
61-70	36951	6603	17.87%	1.154	1.121, 1.187	<.0001
51-60	23809	3909	16.42%	1.003	0.971, 1.037	.8488
41-50	7099	891	12.55%	0.753	0.707, 0.802	<.0001
31-40	1776	130	7.32%	0.442	0.375, 0.522	<.0001
21-30	888	37	4.17%	0.253	0.184, 0.346	<.0001

RCR, rotator cuff repair; RR, relative risk; CI, confidence interval.

P values below .05 are bold.

Figure 1.

Bar graph representing retear percentage by age group.

Figure 2.

Forest plot representing relative risk with 95% confidence intervals for morbid condition (postpropensity score matching) and age group. The solid vertical line indicates a relative risk of 1, which indicates no significant risk. A 95% confidence interval overlapping this vertical line indicates a nonsignificant result. Those to the left of the line denote lower risk, while those to the right denote higher risk.

Patients aged 41-50 showed a significant decrease in risk of retear (RR 0.753, P < .0001, 95% CI 0.707, 0.802) with a retear rate of 12.5%. Patients aged 31-40 had a retear rate of 7.3% with a significant reduction in risk of retear (RR 0.442, P < .0001, 95% CI 0.375, 0.522). Patients aged 21-30 experienced significant reduction in risk of retear (RR 0.253, P < .0001, 95% CI 0.184, 0.347) with a retear rate of 4.1%.

Risk of retear by morbidity

All morbidities resulted in a significant risk increase for rotator cuff retear prior to propensity score matches. (Table III) Postpropensity score matching, no significant difference was found in risk of retears for patients with CIHD (RR 1.005, P = .817, 95% CI 0.961, 1.052), T2DM (RR 0.976, P = .235, 95% CI 0.937, 1.016), and CKD (RR 0.998, P = .955, 95% CI 0.941, 1.060). Patients with smoking history experienced the greatest increase in risk for retear (RR 1.217, P < .0001, 95% CI 1.146, 1.292). Patients with HLD had the second highest relative risk of retear (RR 1.121, P < .0001, 95% CI 1.080, 1.164) followed by patients with HTN (RR 1.118, P < .0001, 95% CI 1.077, 1.159).

Table III.

Relative risk by morbidity: pre and postpropensity score matching.

	Prepropensity score matching					Postpropensity score matching
	RCR (2012-2021)	Retear	RR	95% CI	P value	RCR (2012-2021)	Retear	RR	95% CI	P value
Total	98844	16189
Morbidity
Hypertension	52786	9827	1.348	1.309, 1.387	<.0001	32186	5345	1.118	1.077, 1.159	<.0001
Hyperlipidemia	41147	8145	1.420	1.380, 1.460	<.0001	16717	2817	1.121	1.080, 1.164	<.0001
Chronic ischemic heart disease	15969	3142	1.250	1.207, 1.294	<.0001	16717	2817	1.005	0.961, 1.052	.8169
Type 2 diabetes mellitus	21844	4124	1.205	1.1670, 1.244	<.0001	8732	1322	0.976	0.937, 1.016	.2348
Smoking	9078	1906	1.320	1.264, 1.377	<.0001	5863	1032	1.217	1.146, 1.292	<.0001
Chronic kidney disease	8728	1752	1.253	1.198, 1.310	<.0001	8768	1730	0.998	0.941, 1.060	.9545

RCR, rotator cuff repair; RR, relative risk; CI, confidence interval; CKD, chronic kidney disease; T2DM, type 2 diabetes mellitus; CIHD, chronic ischemic heart failure; HLD, hyperlipidemia; HTN, hypertension.

P values below .05 are bold.

Discussion

Retears following ARCR is a common, but difficult postoperative challenge patients and surgeons must contend with. While previous studies have investigated risk factors associated with the development of retears, this study is one of the largest studies to date evaluating the relationship between medical risk factors and retears following ARCR. Overall, the results of the study found age and numerous medical comorbidities, including HTN, HLD, smoking history, and CKD, to be associated with an increased risk of retear following ARCR.

The results of this study were similar to previously published studies within the literature finding a direct relationship between increasing age and rotator cuff retear following ARCR.^8,28,32 There was an overall retear rate of 16.6%, with the greatest retear rates seen in patients age 81-90. Additionally, there were significant increases in tear rates for all patients over the age of 60. Diebold et al found similar trends, with minimal rates of retear prior to age 50 and a linear increase in subsequent decades.⁹ This trend is in line with current understanding of rotator cuff tears, and the results of this study, as older patients have been shown to present more frequently with full thickness tears than younger patients,^17,29,30 with a direct correlation between increasing tear thickness and diminished postoperative healing potential.^{7,13,16,26,34}

Hypertension is a common risk factor associated with rotator cuff disease. The relationship between hypertension and a rotator cuff tear is in part related to rotator cuff tendon hypovascularity from arterial hypertension induced peripheral hypovascularity, resulting in degeneration of osteotendinous junctions.¹⁰ Within this study, patients with HTN were found to have 1.181 times the risk for retear after propensity matching, a comparable outcome to Gumina et al, although not as significant as their finding of a 2-fold higher risk.¹⁰ In addition to its effects on rotator cuff healing, HTN has been found to be an independent predictor for readmission following ARCR.²⁴ Abnormal blood pressure, cholesterol, and blood sugar have all been implicated in weakened morphologic integrity of the Achilles tendon, bolstering the literature regarding the deleterious effect of these factors on osteotendinous junctions including the rotator cuff.⁶ This study also found a significantly increased risk of retears in diabetic patients. While the impact of diabetes on rotator cuff tears is well studied, the current literature describing the impact of diabetes on retears following ARCR reports mixed results regarding clinical outcomes of diabetics vs. nondiabetics.^4,5,14,22,27 Impaired microcirculation, proinflammatory environments secondary to cytokine overexpression, and altered gene expression are few of the pathogenetic pathways implicated in poor postoperative outcomes recognized in diabetic patients.^14,22 There was no significant relationship found between the presence of T2DM, CKD, and retear rate.

Smoking has been strongly associated with rotator cuff pathogenesis as well as tendon healing. Within this study, smoking was found to increase the risk of retear by 1.113 times following ARCR. In contrast, Kim et al found no correlation between patient smoking history and retear rates.¹⁹ Naimark et al similarly found no difference in retear rates, however with smokers initially presenting with larger tears initially and worse outcome scores and functional improvement in response to ARCR.²⁹ However, as a potent vasoconstrictor, the nicotine within tobacco products can reduce blood supply to the already relative avascular rotator cuff insertion, potentially causing attritional rotator cuff tears with a decreased capacity for healing.^3,28

The relationship between HLD and rotator cuff disease has been extensively studied, with many concluding adverse outcomes in affected patients, including rates of retear.^11,20,35 Excess lipid accumulation within tendons, thought to affect their mechanical properties, have been hypothesized to be the cause.¹² Elevated cholesterol has also been hypothesized to impact the tendons extracellular matrix, lending to injury and poor healing capacity. Abboud and Kim found that total cholesterol, triglycerides and LDL cholesterol concentrations in patients with rotator cuff tendon tears were higher when compared to patients with non-tendon shoulder injuries.¹ Their conclusion was in line with prior research on Achilles tendon rupture and elevated cholesterol as described above. Prior studies have outlined that as patients age, dense connective tissues, such as tendons, can accumulate significant amounts of cholesterol which leads to weakened tendon integrity.

Associations between CKD and CIHD and rotator cuff disease have been seldom studied within the literature. Patients with severe CKD, defined by the requirement of dialysis, were found to have increased risk of postoperative adverse events; in addition to increased operative times and readmission within 30 days of ARCR.²¹ However, no significant relationship between CKD and retear rate was found. Heart disease additionally has been shown to increase risk of rotator cuff tendinopathy, with no studies examining risk of retear following ARCR.² This study found an increased risk for rotator cuff retear following ARCR in patients with CKD, with no significant increase in risk in patients with CIHD.

With the utilization of the TriNetX database, patient data from 63 healthcare organizations across the country was utilized creating a diverse assortment of patient data otherwise unavailable to single institution studies. However, this study has several limitations. Potential confounding factors such as preoperative tear size and thickness were unable to be accounted for in this study. Previous studies have also noted that size of the initial tear is a prominent risk factor for future retear. Due to limitations in data within the TriNetX platform, tear size cannot be clarified and is a limiting factor. Another limitation is that some medical comorbidities are often linked. The results of this study should be interpreted with caution, considering that cardiovascular and kidney pathologies can be interrelated. Additionally, as a result of this study's heterogeneity, the operative procedures were not standardized by surgeons performing the repair, operative technique, or intrinsic factors related to tendon health and healing potential.

Conclusion

Rotator cuff tears are common injuries with retears posing a significant concern in patients receiving ARCR. This study identifies multiple risk factors predisposing to retears after ARCR, many of which are modifiable and can be potentially addressed pre and postoperatively to potentially mitigate the risk of retear. Understanding these potential risk factors for retear may be helpful in optimizing outcomes and guiding surgical treatment for rotator cuff tears.

Disclaimers:

Funding: No funding was disclosed by the authors.

Conflicts of interest: The authors, their immediate families, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.